The overall goal is to define the physiologic, cellular, and molecular mechanisms targeted by these bioactive food components as potential anti-angiogenic agents for colon cancer prevention. The current proposal will develop and characterize the role of angiogenesis in colorectal cancer using transgenic mouse models to understand how specific bioactive food components exhibit their anti-angiogenic activities during colon carcinogenesis. Preventing dysplastic cell masses from gaining vascular access is a critical step in preventing the growth, development and transformation of these abnormal cells to a carcinogenic phenotype. ApcMin/+ mice and the same in mice harboring a thrombospondin-1 null mutation will be used to monitor the rate of growth and development of spontaneous colorectal tumors. Thrombospondin-1 is a potent endogenous angiogenesis inhibitor that is known to suppress tumorigenesis in a p53 null background and in the context of certain carcinogens. Its expression is frequently lost during carcinogenesis in the human colon. Therefore, we expect that ApcMin/+ mice will develop tumors sooner in the thrombospondin-1 null background. Conversely, the transcription factor AhR was recently shown to positively regulate the thrombospondin-1 promoter, and AhR is a known target of the potentially chemopreventive food components resveratrol and circumin. Therefore, we propose that the chemopreventive activities of these compounds may differ in mice lacking thrombospondin-1. This collaboration will test this hypothesis and establish a mouse tumorigenesis model that can be utilized for additional prevention studies using dietary and nondietary agents. The effects of various anti-angiogenic bioactive food components on the phenotypes of colon tumor cells will be examined by measuring the rate of growth, size of tumors, and extent of tumor vascularity. Specific morphologic changes such as vessel permeability and vessel structure will be examined in tissue sections using histology along with changes in expression patterns of VEGF, HIF-1 and its target genes. Effects of dietary supplements on thrombospondin-1 expression in colon epithelia will also be assessed in ApcMin/+ mice by real time PCR. Specific Aims: 1. To examine the anti-angiogenic effects of curcumin and resveratrol individually on proliferation, migration, and tube formation in stimulated human umbilical vein endothelial cells (HUVEC) and the role of altered thrombospondin-1 expression in this activity. 2. To compare the anti-angiogenic effects of curcumin and resveratrol individually on colon cancer in pro-angiogenic ApcMin/+ mice expressing thrombospondin-1 or with a thrombospondin-null mutation. Experiment 1 1. In vitro angiogenesis studies &#9679;Cultured HUVEC will be pre-treated with angiogenic stimulators (i.e., VEGF, FGF2, IL-1 alpha, etc.) before treatment with curcumin and resveratrol, individually. Thrombospondin-1 expression will be assessed by real time PCR. Proliferation, migration, and capillary tube formation will be assessed by standard methods. &#9679;Similarly, cultured primary murine endothelial cells from wild type and thrombospondin-1 null mice will be cultured with curcumin and resveratrol in the presence of angiogenic stimulators, and angiogenic responses will be quantified. &#9679;VEGF and HIF-1 expression will be measured in matching similarly stimulated and treated HUVEC along with cell proliferation. Experiment 2 1. Dietary supplementation &#9679;20 wild type ApcMin/+ transgenic mice and the same harboring the thrombospondin-null mutation at 4 weeks of age will be switched to the AIN-76A diet supplemented with either daily 1g/kg oral curcumin (10 mice;Tx for 12 weeks) or fed the AIN-76A control diet with 0.01% resveratrol in the drinking water also containing 0.4% ethanol (10 mice;Tx for 12 weeks) and maintained on these diets throughout the study. &#9679;10 wild type ApcMin transgenic mice harboring the thrombospondin-null mutation will be maintained on a characterized mice diet (or AIN-76A diet) until sacrifice. 2. Sample collection &#9679;After completing their dietary interventions, 3-5 mice from each of the two dietary treatments and 3-5 control mice will receive i.p. BrdU one hour before prior to euthanasia for colonic epithelial cell cytokinetic analyses. &#9679;After completing their control or dietary interventions, the remaining animals will be sacrificed. All mice will be euthanized by CO2 affixation, their GI tracts removed and tumor burden will be determined. &#9679;The gastrointestinal tract, blood, urine of these animals will be collected and be analyzed for changes in VEGF and HIF pro-angiogenic responses to control and dietary treatments. &#9679;Colonic tumor number, size and volume will be measured for each of the control, curcumin and resveratrol treatment animal groups. Future Studies Specific Aims: Aim 1: Examine the effects of various anti-angiogenic dietary components on the phenotypes of colon tumor cells. We will follow the effects of anti-angiogenic bioactive food components on the surrounding colonic microenvironment impacting the development and growth of colon tumors. Aim 2: Define the effective and physiologically achievable concentrations of those dietary components using the data from Aim 1. Aim 3: Examine the altered balance between pro- and anti-angiogenic mediators that are secreted at different time points and the molecular changes occurring at the level of signal transduction, transcription or translation. Microarray and/or proteomic analysis will be used to determine these molecular changes. Aim 4: Identify the specific biomarkers for dietary components that influence the angiogenic process as well as the colon cancer risk or incidence. These biomarkers could be molecular targets that could be beneficial for intervention, or correlative to the identified genes/proteins by microarrays or proteomic screening.